This invention relates to electrochemical capacitors and especially to solid state capacitors having a plurality of electrodes and electrolyte layers.
Electrochemical capacitors fall into two broad categories; (1) double layer capacitors which rely solely on interfacial charge separation across the electrical double layer and (2) pseudocapacitors which have enhanced charge storage derived mainly from Faradaic transfer parallel with the double layer. Both types of capacitor offer major increases in capacitance density and energy storage capability when compared with conventional dielectric capacitors. They have the potential for extremely low values of electrical leakage unsurpassed by any other capacitive energy storage device. Electrochemical capacitors bridge the energy and power gap between batteries and capacitors. At the same time, they can exhibit both the cycle life and the stability expected of passive components.
In double layer capacitors, the components of the double layer capacitance are in electrical series leading to the equation EQU 1/C.sub.Total =1/C.sub.H +1/C.sub.D
where C.sub.H and C.sub.D are the Helmholtz and diffuse layer capacitances respectively. If C.sub.D is large, as for example at high electrolyte concentrations, then the effective capacitance of the interface is more nearly equal to the Helmholtz capacitance. Conversely, when C.sub.D is low, as it will be at low concentrations, then the total capacitance tends to this value. Since concentrated electrolytes are utilized in electrochemical capacitors in order to achieve minimum resistance or ESR, the original Helmholtz concept will generally yield the more accurate values of capacitance. Since the charge or discharge of the double layer involves only a dipole reorientation process, it is a very rapid electrode reaction. When, on the other hand, charge storage is dependent on a charge transfer process, the kinetics will tend to be slower, and can be governed by diffusion in instances where chemical species from the bulk of either the electrode or the electrolyte are involved. Charge cannot leak across the double layer except by a charge transfer process. As a result, electrical leakage in double layer capacitors is intrinsically absent.
Pseudocapacitance is in parallel with the components of true double layer capacitance and thus gives rise to the possibility of considerably enhanced charge storage. Electrical response of a pseudocapacitor can often closely resemble that of a conventional double layer capacitor over a wide range of potential between the cathodic and anodic limits of electrolyte decomposition. The apparent DC capacitance (really the true double layer capacitance plus the pseudocapacitance) can be derived from triangular sweep/voltage curves by the equation EQU C=idt/dV
Also, the integral capacitance can be readily determined by means of constant current charging curves. The materials used in a pseudocapacitive couple include electrode materials which can be prepared in a stable, high surface area form and materials which can undergo reversible surface and or near-surface charge transfer reactions at relatively constant voltage. Certain of the conducting transition metal oxides, notably RuO.sub.2, and IrO.sub.2 have been found to fulfill these requirements. The clear advantage inherent in the use of a pseudocapacitive material for high rate, high energy density electrochemical capacitors is in the significantly higher capacitance densities achievable when compared with those for double layer capacitors.
The main design parameters to be considered in the selection of an electrochemical energy storage capacitor are voltage, capacitance, density, ESR, leakage current, and energy density. Other significant parameters are the load and charge-back profiles, duty cycle and repetition rate. The electrical means of coupling the capacitor with the load must also be defined.
The energy stored in a capacitor is given by the simple formula EQU 1/2CV.sup.2
while the basic equations relating voltage and time for the charge and discharge of a capacitor which in turn govern power output are EQU V.sub.t =V.sub.o (1-e.sup.-t/Rc) (charge) EQU V.sub.t =V.sub.o e.sup.-t/RC (discharge)
where R is the equivalent series resistance (ESR). Maximum current and power levels are achieved by devices possessing the lowest value of ESR and the highest voltages. For constant coulombs charged into a device, the energy stored is directly proportional to voltage as a consequence of the fact that the expression 1/2 CV.sup.2 is equivalent to 1/2 QV.
Solid state electrochemical capacitors can be made by using solid state electrolytes in place of more conventional liquid electrolyte materials. The advantages sought in solid state capacitors include the elimination of electrolyte leakage, improved operation at high and low temperatures, elimination of the need for a discrete separator element, the ease of miniaturization, multiple shape factors and the possibility of simple reliable designs for bipolar electrode cell stacks to allow fabrication of high voltage devices. However, the major problem with solid electrolytes is that they have much higher electrical resistances than liquid electrolytes which results in relatively high values of ESR for solid state capacitors. Typically, the conductivity of liquid electrolytes exceeds that of solid electrolytes by several orders of magnitude. Such low conductivity means that although the capacitor may have the ability to store a great deal of energy per unit volume, the time to discharge the energy stored in the capacitor is quite long and may make the capacitor unsuitable for many electronic applications for which the stored energy must be supplied at high voltages in milliseconds rather than in seconds or minutes. Since high voltages require multi-layer designs with cells connected in series, even a very low ESR for a cell can amount to an unacceptable ESR for the entire device. One example of such an application is the implantable heart defibrillator where high voltages (e.g. 800 volts) must be delivered from a capacitor within 5-15 milliseconds.
The conductivity issue has been addressed by the development of improved conductive polymeric electrolyte materials in both capacitors and batteries. For example, in U.S. Pat. Nos. 4,618,407 and 4,748,542 to Lundsgaard a polymer (e.g. polyethylene oxide), rendered ionically conductive by the addition of an inorganic salt (e.g. LiClO.sub.4, NaClO.sub.4, LiCF.sub.3 SO.sub.3, or LiBF.sub.4) was combined with activated carbon-containing electrodes to produce a multi-layer solid-state electrochemical capacitor. In U.S. Pat. No. 4,830,939 issued to Lee et al., a solid electrolyte for electrochemical cells is formed by mixing a liquid polymerizable compound, a radiation inert ionically conducting liquid and an ionizable alkaline metal salt and subjecting them to polymerizing radiation. In Abraham et al., Li.sup.+ -Conductive Solid Polymer Electrolytes with Liquid-Like Conductivity, J. Electrochem. Soc., Vol 137, No. 5, May 1990, a polymer network of polyacrylonitrile, poly(tetraethylene glycol diacrylate) or poly(vinyl pyrrolidone) is disclosed to immobilize a liquid solvent such as ethylene carbonate or propylene carbonate and a lithium salt. In a paper by Abraham, Room Temperature Polymer Electrolyte Batteries Fourth international Rechargeable Battery Seminar, Florida (March 1992), additional polymer networks using poly(bis-((methoxy ethoxy) ethoxy) phosphazene), poly(propylene oxide), and poly(ethylene oxide) are disclosed. In Kanbara et al., New solid-state electric double-layer capacitor using poly (vinyl alcohol)-based polymer solid electrolyte, Journal of Power Sources 36 (1991) 87-93 PVA was disclosed to dissolve large amounts of lithium salts to provide a solid-state electrolyte with good conductivity for double layer capacitors. In U.S. Pat. No. 4,792,504 issued to Schwab et al., a solid polymer electrolyte is disclosed with a continuous network of polyethylene oxide containing a dipolar aprotic solvent and a metal salt.
Another component of electrochemical capacitors that has received attention is the electrode. Carbon electrodes have been studied extensively. Carbon black has the drawback of high resistivity arising out of poor particle-to-particle contact. However, since the energy density of the capacitor is directly proportional to accessible surface area of electrode materials and since carbon black is known as an electrode material which can have surface areas in the range of 1000-2000 m.sup.2 /g, it is still highly desirable for electrochemical capacitor applications. The conductivity of solid state electrochemical capacitor electrodes can be enhanced by adding to the carbon a sulfuric acid aqueous solution such as that used in double layer capacitors sold under the tradename SUPERCAP by Nippon Electric Company (NEC). Conductivity can also be improved by including in the carbon electrode an ionically conducting polymer (e.g. polyethylene oxide) and an inorganic salt as set forth in U.S. Pat. Nos. 4,618,407 and 4,748,542 to Lundsgaard. Also, in Japanese Kokai Patent Application No. HEI 2[1990]-39513, carbon is impregnated with a polymeric solid electrolyte made with a solid solution of a polyether polymer and an alkali metal salt. However, carbon electrodes using these solid electrolytes still lack the conductivity needed for many double layer capacitor applications.
It is therefore an object of the present invention to provide an electrochemical capacitor with improved ESR.
It is also an object of the present invention to provide a high voltage, multi-layer electrochemical capacitor with an improved ESR.